Hi Cyril,
"glad" I am not the only user with these problems ;-)
The model fitting looks appropriate (the GLM is indeed a wonderful
instrument LOL). But the data are really odd! Is this the whitening
filter????
Cheers,
Bruno
On 02/08/2016 14:17, PERNET Cyril wrote:
> Hey Bruno,
>
>
> no proper testing but indeed adding the 2nd derivatives also tends to
> give me odd results .. but I haven't investigated why
>
>
> cyril
>
>
>
> ------------------------------------------------------------------------
> *From:* SPM (Statistical Parametric Mapping) <[log in to unmask]> on
> behalf of Bruno L. Giordano <[log in to unmask]>
> *Sent:* 02 August 2016 14:20:33
> *To:* [log in to unmask]
> *Subject:* [SPM] Failure of dispersion derivative?
>
> Hello,
>
> I have quite some event-related data that would be nice for fitting a
> model with the HRF, temporal and dispersion derivatives (ISI = ~4 s; TR
> = 1).
>
> However, when adding to the regressors also the dispersion derivative, I
> am getting quite odd results (see attached figure; same single-subject
> data, same auditory cortex voxel - stimuli are sounds, different number
> of derivatives).
>
> I suspect this apparent "failure" (I wouldn't really expect a negative
> BOLD response in this auditory voxel) has to do with the drastic change
> in the model residual when the dispersion derivative is added, which, in
> turn, changes drastically the whitening filters and the whitened data.
>
> I also suspect it might be a good idea to drop (reluctantly) the
> dispersion derivative. Does anyone on this list have experience with
> similar problems?
>
> Thank you,
>
> Bruno
>
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Bruno L. Giordano, PhD
> Institute of Neuroscience and Psychology
> 58 Hillhead Street, University of Glasgow
> Glasgow, G12 8QB, Scotland
> T +44 (0) 141 330 5484
> Www: http://www.brunolgiordano.net
> Email charter: http://www.emailcharter.org/
>
>
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Bruno L. Giordano, PhD
Institute of Neuroscience and Psychology
58 Hillhead Street, University of Glasgow
Glasgow, G12 8QB, Scotland
T +44 (0) 141 330 5484
Www: http://www.brunolgiordano.net
Email charter: http://www.emailcharter.org/
|